Skip to main content
POST
/
proxy
/
vidu
/
start-end2video
cURL
curl --request POST \
  --url https://api.comfy.org/proxy/vidu/start-end2video \
  --header 'Content-Type: application/json' \
  --data '
{
  "aspect_ratio": "<string>",
  "audio": true,
  "audio_type": "all",
  "bgm": true,
  "callback_url": "<string>",
  "duration": 123,
  "enhance": true,
  "images": [
    "<string>"
  ],
  "is_rec": true,
  "meta_data": "<string>",
  "model": "<string>",
  "movement_amplitude": "auto",
  "off_peak": true,
  "payload": "<string>",
  "priority": 123,
  "prompt": "<string>",
  "resolution": "<string>",
  "seed": 123,
  "style": "general",
  "voice_id": "<string>",
  "watermark": true,
  "wm_position": 123,
  "wm_url": "<string>"
}
'
{
  "credits": 123,
  "state": "created",
  "task_id": "<string>",
  "aspect_ratio": "<string>",
  "bgm": true,
  "created_at": "2023-11-07T05:31:56Z",
  "duration": 123,
  "images": [
    "<string>"
  ],
  "model": "<string>",
  "movement_amplitude": "auto",
  "off_peak": true,
  "payload": "<string>",
  "prompt": "<string>",
  "resolution": "<string>",
  "seed": 123,
  "style": "general",
  "watermark": true
}

Body

application/json
​
aspect_ratio
string
​
audio
boolean

Enable direct audio-video generation capability (default true for q3 model)

​
audio_type
enum<string>

Audio type when audio is true: all (sound effects + vocals), speech_only, sound_effect_only. Ineffective for q3 model

Available options:
all,
speech_only,
sound_effect_only
​
bgm
boolean

Add background music to generated video (ineffective for q3 model)

​
callback_url
string

Callback URL for task status updates

​
duration
integer<int32>

Video duration in seconds. viduq3-pro: 1-16, viduq2-pro-fast: 1-10, viduq2-pro/turbo: 1-8

​
enhance
boolean
​
images
string[]

Images for img2video (accepts 1 image as start frame)

​
is_rec
boolean

Use recommended prompt (consumes additional 10 credits)

​
meta_data
string

Metadata identification, JSON format string for custom metadata

​
model
string

Model name: viduq3-pro, viduq2-pro-fast, viduq2-pro, viduq2-turbo, viduq1, viduq1-classic, vidu2.0

​
movement_amplitude
enum<string>

Movement amplitude of objects in frame (ineffective for q2, q3 models)

Available options:
auto,
small,
medium,
large
​
off_peak
boolean

Off peak mode (lower cost, tasks generated within 48 hours)

​
payload
string

Transparent transmission parameters (max 1048576 characters)

​
priority
integer<int32>
​
prompt
string

Text prompt for video generation (max 2000 characters)

​
resolution
string

Resolution: 360p, 540p, 720p, 1080p, 2K (availability depends on model and duration)

​
seed
integer<int32>

Random seed (defaults to random if not specified)

​
style
enum<string>
Available options:
general,
anime
​
voice_id
string

Voice ID for audio (ineffective for q3 model)

​
watermark
boolean

Add watermark to video (default false)

​
wm_position
integer<int32>

Watermark position: 1 (top left), 2 (top right), 3 (bottom right, default), 4 (bottom left)

​
wm_url
string

Watermark image URL (uses default watermark if not provided)

Response

OK

​
credits
integer<int32>
required
​
state
enum<string>
required
Available options:
created,
processing,
queueing,
success,
failed
​
task_id
string
required
​
aspect_ratio
string
​
bgm
boolean

Whether background music was added

​
created_at
string<date-time>
​
duration
integer<int32>
​
images
string[]
​
model
string
​
movement_amplitude
enum<string>
Available options:
auto,
small,
medium,
large
​
off_peak
boolean

Off peak mode status

​
payload
string

Transparent transmission parameters

​
prompt
string
​
resolution
string
​
seed
integer<int32>
​
style
enum<string>
Available options:
general,
anime
​
watermark
boolean

Whether watermark was added